【作者】 潘娜；

【导师】 毛宇光；

【作者基本信息】 南京航空航天大学 ， 计算机软件与理论， 2005， 硕士

【摘要】 不完全信息数据库与 SQL 语言的语义是数据库理论的两个重要研究方向,不仅具有重要的理论价值,而且还具有很高的应用价值。本文借助中介逻辑谓词演算系统 MFM,对不完全信息系统的理论和应用进行了研究。为了更好地处理不完全信息数据库的查询,在对目前数据库处理空值所存在的问题进行分析的基础上,对 SQL 查询进行了修改和扩充,构造了针对不完全信息数据库的查询语言,使之能够比较正确地处理不完全信息。同时,本文将 SQL 查询结果分为确定信息和可能信息两种类型,并针对产生不同结果的 SQL 查询进行描述,并且在 Linux系统下,通过使用 Yacc 语法分析工具和 Lex 词法分析工具实现了不完全信息数据库基本的查询功能。其次,以 MFM 为理论基础,构造了一个针对可能信息的SQL 查询的形式的三值元组关系演算 MFPC,在对此演算进行构造的基础上通过语法转化规则把 SQL 查询转化为 MFPC,从而完整地定义了 SQL 查询的形式语义。最后,建立了中介逻辑谓词演算系统 MFM的表推演系统,并证明了该系统的可靠性和完备性,该系统可作为不完全信息的自动推演系统。更多还原

【Abstract】 Incomplete information database and formal semantics of SQL are two important fields in database theory. They not only have important theoretical values, but also have great practical values. This paper investigates the theory and application of incomplete information system by using the medium logic predicate calculus system MFM. In this paper, the queries of SQL are modified and the query language to the incomplete information databases is founded. The results of SQL queries are classified into two types: sure information and maybe information. And the descriptions of SQL query are given in allusion to the different types of the answer. And the query that can deal with the incomplete information database is implemented in Linux system by using the tools of Yacc and Lex. A formal three-valued predicate calculus MFPC is constructed, which is based on the medium logic predicate calculus system MFM, and then the SQL query sentences are transformed into MFPC by syntax transforming rules. The formal semantics of SQL query is completely defined by this means. At last, the tableau system of medium logic predicate calculus system MFM is constructed, and the soundness and completeness of the system is proved. This system can act as the automated reasoning system in incomplete information environments.更多还原